Collaborative Research: Decoding the Corrosion of Borate Glasses: From Fundamental Science to Quantitative Structure-Property Relationships

合作研究：解码硼酸盐玻璃的腐蚀：从基础科学到定量结构-性能关系

• 批准号: 2034856
• 负责人: Aditya Kumar
• 金额: $ 26万
• 依托单位: Missouri University of Science and Technology
• 依托单位国家: 美国
• 项目类别: Continuing Grant
• 财政年份: 2021
• 资助国家: 美国
• 起止时间: 2021-04-01 至 2025-03-31
• 项目状态: 未结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=2034856&HistoricalAwards=false
• 关键词: Collaborative; Research; Decoding; Corrosion; Borate

NON-TECHNICAL DESCRIPTION: Chemical durability of glass is a topic of interest today; fundamental understanding is of paramount importance to the glass industry and to the pursuit of overcoming various challenges relevant to the well-being of humanity and the environment, including nuclear waste management and development of novel biomaterials. This project aims at understanding the fundamental science governing corrosion of multicomponent borate glasses, achieved through the unification of experimental studies and artificial intelligence. Successful completion of this project is expected to lay the foundation of new fundamental knowledge to understand and describe composition-structure-property relationships in glass corrosion, and advance new machine learning-based models to promptly and reliably predict the corrosion behavior of borate glasses. The U.S. glass/materials industry is facing a severe shortage of experienced glass engineers/scientists. The project reduces this shortage by training undergraduate and graduate students in glass science and engineering, thus providing a talent pool for the U.S. glass/materials industry, academia, and national laboratories. The education and outreach activities are designed to invoke interest in students and teachers at the middle and high school levels, in addition to the training of undergraduate and graduate science and engineering students. TECHNICAL DETAILS: Our current understanding of glass corrosion is based primarily on empirical data, as there is still no complete consensus on the primary mechanism of glass dissolution that applies across a wide composition space. Therefore, there is an exigent need to develop robust, fundamental understanding of the linkage(s) between chemical composition, atomic/molecular structure, and chemical durability of glasses in order to address crucial and scientifically challenging problems (e.g., designing glasses with desired chemical durability). Accordingly, the project aims at combining the strengths of experimental studies and artificial intelligence to reveal the underlying mechanisms that dictate the dissolution behavior of borate glasses in aqueous environments; and developing a cloud-based quantitative structure-property relationship (QSPR) model – powered by theory-guided machine learning engine – to predict the time-dependent corrosion behavior of oxide glasses. Enabling the materials-by-design approach – which is in alignment with the U.S. Materials Genome Initiative – this project is a pioneering effort, representing a leap forward in designing oxide glasses with controlled chemical durability. Apart from revealing fundamental drivers of glass corrosion and advancing a QSPR model to reliably predict glass corrosion, a significant outcome of the project is the development of a talent pipeline of undergraduate and graduate students well-trained in glass/materials science and machine learning. Further, the project's education plan incorporates a foundational, spiral approach that builds interest at the elementary, middle, and high school level students.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

非技术描述：玻璃的化学耐久性是当今人们感兴趣的话题；基本的理解对于玻璃工业和追求克服与人类福祉和环境相关的各种挑战至关重要，包括核废料管理和新型生物材料的开发。该项目旨在了解控制多组分硼酸玻璃腐蚀的基础科学，通过实验研究和人工智能的统一来实现。该项目的成功完成有望为理解和描述玻璃腐蚀中的成分-结构-性能关系奠定新的基础知识，并提出新的基于机器学习的模型，以及时可靠地预测硼酸盐玻璃的腐蚀行为。美国玻璃/材料行业正面临着经验丰富的玻璃工程师/科学家的严重短缺。该项目通过培养玻璃科学和工程专业的本科生和研究生，从而为美国玻璃/材料行业、学术界和国家实验室提供人才库，从而减少了这一短缺。教育和推广活动的目的是唤起初中和高中学生和教师的兴趣，以及培训本科和研究生科学和工程专业的学生。技术细节：我们目前对玻璃腐蚀的理解主要是基于经验数据，因为在玻璃溶解的主要机制上仍然没有完全的共识，适用于广泛的成分空间。因此，迫切需要对玻璃的化学成分、原子/分子结构和化学耐久性之间的联系进行强有力的、基本的理解，以解决关键的、具有科学挑战性的问题（例如，设计具有理想化学耐久性的玻璃）。因此，该项目旨在结合实验研究和人工智能的优势，揭示决定硼酸盐玻璃在水环境中溶解行为的潜在机制；并开发基于云的定量结构-性能关系（QSPR）模型-由理论指导的机器学习引擎驱动-预测氧化玻璃的随时间腐蚀行为。实现材料设计方法-这与美国材料基因组计划一致-该项目是一项开创性的努力，代表了设计具有可控化学耐久性的氧化玻璃的飞跃。除了揭示玻璃腐蚀的基本驱动因素和推进QSPR模型以可靠地预测玻璃腐蚀之外，该项目的一个重要成果是培养了一支在玻璃/材料科学和机器学习方面受过良好训练的本科生和研究生的人才队伍。此外，该项目的教育计划采用了一种基础的、螺旋式的方法，在小学、初中和高中阶段培养学生的兴趣。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

Predicting Dissolution Kinetics of Tricalcium Silicate Using Deep Learning and Analytical Models

• DOI: 10.3390/a16010007
• 发表时间: 2022-12
• 期刊: Algorithms
• 影响因子: 2.3
• 作者: Taihao Han;Sai Akshay Ponduru;Arianit A. Reka;Jie Huang;G. Sant;Aditya Kumar

A Deep Learning Approach to Design and Discover Sustainable Cementitious Binders: Strategies to Learn From Small Databases and Develop Closed-form Analytical Models

• DOI: 10.3389/fmats.2021.796476
• 发表时间: 2022-01
• 作者: Taihao Han;Sai Akshay Ponduru;R. Cook;Jie Huang;G. Sant;Aditya Kumar

Deep learning to predict the hydration and performance of fly ash-containing cementitious binders

• DOI: 10.1016/j.cemconres.2023.107093
• 发表时间: 2023-03
• 期刊: Cement and Concrete Research
• 影响因子: 11.4
• 作者: Taihao Han;Rohan Bhat;Sai Akshay Ponduru;A. Sarkar;Jie Huang;G. Sant;Hongyan Ma;N. Neithalath;Aditya Kumar

Machine Learning Enabled Models to Predict Sulfur Solubility in Nuclear Waste Glasses

机器学习模型可预测核废料玻璃中的硫溶解度

• DOI: 10.1021/acsami.1c10359
• 发表时间: 2021
• 期刊: ACS Applied Materials & Interfaces
• 影响因子: 9.5
• 作者: Xu, Xinyi;Han, Taihao;Huang, Jie;Kruger, Albert A.;Kumar, Aditya;Goel, Ashutosh
• 通讯作者: Goel, Ashutosh

Machine learning enabled closed‐form models to predict strength of alkali‐activated systems

机器学习使封闭式模型能够预测碱激活系统的强度

• DOI: 10.1111/jace.18399
• 发表时间: 2022
• 期刊: Journal of the American Ceramic Society
• 影响因子: 3.9
• 作者: Han, Taihao;Gomaa, Eslam;Gheni, Ahmed;Huang, Jie;ElGawady, Mohamed;Kumar, Aditya
• 通讯作者: Kumar, Aditya